Amplied Doppler shift observed in diraction images as function of the COBE ether drift direction

Carlos. E. Navia and Carlos. R. A. Augusto

Instituto de F sica Universidade Federal Fluminense, 24210-130, Niteri, RJ, Brazil o (Dated: April 7, 2006) We report results on an one-way light path laser diraction experiment as a function of the laser beam alignment relative to the Earths velocity vector obtained by COBE measurements of the Doppler shift in the cosmic microwave background radiation (CMBR). An amplied Doppler shift is observed in the diraction images, and the eect is compatible with a dipole speed of light anisotropy due to Earths motion relative to the CMBR rest frame, with an amplitude of c/ = 0.00123. This amplitude coincides with the value of the dipole temperature anisotropy c T /T = 0.00123 of the CMBR obtained by COBE. Our results point out that it is not possible to neglect the preferred frame imposed by the cosmology and they are well described by the Ether Gauge Theory (an extension of the Lorentzs ether theory) and it satises the cosmological time boundary condition.

arXiv:astro-ph/0604145 v1 6 Apr 2006

1.

INTRODUCTION

The Special Relativity Theory (SRT) introduces an universal fundamental constant, the speed of light in space, it is constant in all frames. Hence all reference frames are completely equivalent and an Ether rest frame is superuous. The experimental basis of the SRT has been pointed out by T. Roberts [1], where a lot of experimental tests done by many authors are summarized. Following this article we found several experiments about round-trip test of light speed isotropy (two-way light path), from the traditional Michelson-Morley to modern Laser/Maser tests, as well as, about one-way light path test of light speed isotropy. In all cases, the experiments have been mounted in Earths surface on a rotating table or xed to the Earth, in this last case it is looked for eects due to the Earths rotation. Only the upper limits have been found and they are in agreement with a null eect, all results conrm the prediction of the SRT, or in other words, the SRT has not been refuted by any experiment. However, recently there are evidences suggesting that the propagation of light over cosmological distances has anisotropy characteristics[2], this means that the speed of light is not a true constant, it depends on direction and polarization. This is a further indication in favor of the existence of a preferred reference frame. This picture is in agreement with the interpretations of the COBE[3] measurements giving the Earths absolute velocity in relation to the uniform cosmic microwave background radiation(CMBR). Of course, there are also interpretations claiming that the COBE measurements only give a velocity for the relative motion between the Earth and the CMBR [4]. For instance, it is possible to remove the Earth motion to obtain a virtual image, where an isotropic distribution of CMBR with small uctuations (T /T 105 ) can be seen. On the other hand, the Big Bang cosmology is well described by the Robertson-Walker-Friedmann metric[5], the time in this metric called as the cosmological time has an absolute beginning and it coincides with the begin of the Big Bang, in other word, it is possible to estimate the absolute age of the universe. In short, the COBE measurements of the Earth motion relative to the CMBR rest frame on an inationary scenario, described by the RobertsonWalker-Friedmann metric, lead to the concept of absolute space-time, because in this scenario it is not possible to remove the Earths motion without altering the Robertson-Walker-Friedmann metric[6]. The consequences of an absolute space-time assumption are the breakdown of the symmetry-energy and symmetrymomentum (linear and angular) conservation laws. The extension of the eects of the violation of these symmetries is still not known. So far, we know that probably they are relevant only in cosmological scale. In fact, in 1999, Coleman and Glashow began to wonder, if Lorentzs symmetry is certainly an exact symmetry of the nature, in the region close to the Placks scale energy. The Lorentz symmetry violation may be correlated to the existence of an absolute reference frame created by the Big-Bang. The main objective of this paper is to present experimental results on an one-way light path laser diraction experiment mounted in the shell of the TUPI muon telescope [7] and that shows clearly that the speed of light depends on the propagation direction. The eect is observed as a amplied Doppler shift in the diraction images as function of the laser beam alignment relative to the Earths velocity vector. The experiment is an improved version about light diraction suggested by C. M. G. Lattes [8] in 1980. The results are well described by the the so called Ether Gauge Theory (EGT), and it is an extension of the Lorentzs ether theory. The gauge of the length is: the moving rods undergo the shortening by Lorentz factor and the gauge of the time is: the moving clocks undergo the slowdown by Lorentz factor. The EGT scenario satises the cosmological time boundary condition (see next section) and leads naturally to a non-isotropic one-way light propagation relative to

2 the Earth frame and where there is a preferred direction. The assumption of a preferred frame agree also with a previous analysis made by Brans and Stewart [9] where a description of the topology of the universe has imposed a preferred state of rest so that the principle of special relativity, although locally valid, is not globally applicable. The analysis of The Global Positioning System (GPS) carried out by Hatch [10] provides also strong indirect evidence for the presence of an ether-drift velocity.
2. ETHER GAUGE THEORY FOR LIGHT PROPAGATION

We presents here a straightforward analysis about light propagation on the basis of the Ether Gauge Theory (EGT) formalism based upon a modication of the Lorentz ether theory in order to include the cosmological time constraint. This exigency is satised by Tangherlini [11] and Selleri [12] transformation. This scenario admit an absolute referential frame H called hereafter as the Hubbles frame [8] and it is linked with the CMBR rest frame. The Hubbles frame is dened as: (a) The frame where the CMBR is isotropic and (b) the frame where light propagation is isotropic. For simplicity and without loss of generality, only a bi-dimensional analysis is made. The Tangherlini-Selleri transformation from H to E in motion relative to H with velocity parameter and where we choose the xH and xE as the direction of the relative motion has been found to be xE = (xH V tH ), yE = yH and tE = 1 tH ,
2

(1)

where = (1 ). The main dierence between the EGT transformation (Tangherlini-Selleri transformation) and the SRT transformation (Lorentz transformation), constitutes the transformation of the time, the Tangherlini-Selleri time transformation satised the cosmological constraint (synchronization of clocks), if tH = 0 then tE = 0. The main consequence of this scenario is that, in the E frame, the axis-x (direction of V ) behaves as a preferred direction. The transformation equations for the velocity of a moving point can be found dierentiating Eq.(1) with respect to tE . dxE dyE = VEx = 2 (VH cos H V ), = VEY = (VH sin H ). dtE dtE Now, if we denote cEx = VEx , cEy = VEy and cH = VH , the speed of light components in the E frame are cEX = cH 2 (cos H ), and the module of this velocity can be obtained as cE = c2 + c2 = cH 2 (1 cos H ). EX EY (4) cEY = cH sin H , (3) (2)

The last equation means that in the E frame the speed of light is not a constant, but it depends on H . Fig.1 shows the space-time diagram (light cone) in the E frame. It is possible to see that the axis-X (direction of V = VE ) behaves as a preferred direction. The cone of light is symmetric in relation to the Y-axis, while it is asymmetric in relation to the X-axis. The slope of the lines is proportional to the inverse value of the speed of light. For illustrative purpose, a large value for (= 0.3) in this diagram has been used. The eect, for instance, due to the Earth motion relative to the CMBR rest frame is tiny ( = 0.001237). From this picture, it is possible to see that there is a resulting vector speed of light, and it is represented as Ve in Fig.1, and hereafter it will be called as the emission vector. The emission vector points in opposite direction of the vector VE and, following Eq.(4), the emission vector has a module given by |Ve | = cH 2 (1 + ) cH 2 (1 ) = 2cH 2 , with amounts to |Ve | 2 |VE |. The eect due to the Earth motion gives |Ve | 742 km s1 . On the other hand, the one-way (forward and backward) speed of light in the Earth frame (E ) are denes as cf = cH 2 [1 cos(H )], E cb E = cH [1 cos(H + )],
2

(5)

(6) (7)

Under this conditions, the one-way speed of light dipole anisotropy is dened as cf cE cE = E , cE cE (8)

3 where cE is the average values as cE = with Eq.(6) and Eq.(7), Eq.(8) becomes cE = cos H , cE (10) cb + cf E E , 2 (9)

Thus the EGT scenario predicts a dipole speed of light anisotropy with an amplitude cE /cE = (= 0.001237), which coincides with the value of the dipole temperature anisotropy (T /T = 0.001237) of the CMBR as observed by COBE. Fig.2 summarized the situation, where the dependence of the dipole speed of light anisotropy is plotted as function of H . Following Fig.2, it is possible to see that in Earth-based experiments, the choice of the direction of the light beam is essential, An random choice for the light beam direction, as in experiment mounted on a rotating table, can easily lead to ambiguous results. In order to compensate the change of direction due to the Earth rotation, the Earth-based experiments must be mounted like a optical telescope using an equatorial ensemble, for intance.
3. THE TUPI LASER DIFFRACTION EXPERIMENT

The TUPI muon telescope is installed on the campus of the Universidade Federal Fluminense, Niteri, Rio de o Janeiro-Brazil at sea level. The position is: latitude: 220 54 33 S, longitude: 430 08 39 W. The TUPI muon telescope has an equatorial assembly and a servo-mechanism which allows the axis of the telescope to be pointed so as to accompany a given source [7, 13, 14]. On the other hand, the bigger temperature anisotropy in the cosmic background radiation is represented by a dipole with an amplitude of T /T = 1.23 103 = 0.123% which arises from the motion of the solar system barycenter with a velocity VE = 371 0.5kms1 ( = 0.001237 0.000002) at 68%CL, relative to the so called CMBR rest frame and towards a point whose equatorial coordinates are (, ) = (11.20h 0.01h , 7.220 0.080 )[15]. This direction points for the Crater constellation. A laser light diraction experiment has been mounted in the shell of the TUPI muon telescope, as is represented in Fig.3. The layout of the diraction device is shows in Fig.4, the experiment permits us to obtain the variations of the diraction line positions, Yn , as a function of the laser beam alignment, H , relative to the Earths velocity vector, obtained by COBE. This experiment is an improved version of an experiment accomplished in 1980 in Campinas Brazil and suggested by C. M. G. Lattes (with unpublished results). Just as in the modern version of the Michelson-Morley experiment [16], where anisotropies are sought in the light (radiation) propagation due the Earth rotation, the Lattes version tried to measure variations of the positions (Yn )(see Fig.4) of the diraction lines due to the Earths rotation. This requires to make measurements over long time of at least several days to weeks, and it is necessary to take into account the atmospheric pressure and temperature variations among others, because the daily atmospheric variations may mask the eect. In the TUPI laser diraction experiment, a complete set of measurements is made in only ten minutes. In addition, just as it is shown below, the Yn variations are bigger when the direction of the laser beam goes from parallel to anti-parallel to the Earths velocity vector (ether drift). Fig.4 shows a schematic diagram of the TUPI laser diraction experiment in the Earth (E frame). Following this gure, it is possible to see that the maxima of intensity of the diraction images satisfy the condition sin n = n E , with (n = 0, 1, 2, ....), (11)

where E is the wave length in the E frame and is the diraction grating step. Under the conditions sin n tan n = we have Yn = d0 tan n n d0 E . (12)

The wave length E can be obtained as the ratio between the speed of light cE and the light frequency E resulting into E = cE /E . An expression for the E as a function of the angle H can be obtained using the transformation equations from E to H supplied by the EGT formalism, which for cE gives cE = cH 2 (1 cos H ). (13)

4 On the other hand, according to the EGT formalism (see section 2), the space-time diagram or light cone in the E frame is asymmetric in relation to the axis parallel to the Earths speed vector, VE , and there is the emissions vector (see Fig.1) represented as Ve in Fig.4, where P represent the point of emission of light in the laser. This vector points in opposite direction of the Earths motion and its module is |Ve | 2 |VE |. This is equivalent to = consider a source moving away at speed Ve = 2 VE . While in the H frame we have a source moving away at speed Ve 2 VE VE = VE . = Consequently, a relation for the Doppler eect must be used to obtain E . The relation for the Doppler eect obtained on the basis of the EGT formalism is the same as the relation obtained by using the SRT one since they are in agreement with experimental measurement [17]. This suggest that locally, both the SRT and the EGT theories are indistinguishable, although they dier when global aspects are considered. The relation of the Doppler eect is E = Including the last two expressions in E result in E = (1 cos H )2 0 , (1 2 )3/2 (15) 1 2 H . (1 cos H ) (14)

where 0 = cH /H . Thus the combined eect, the cE = cE (H ) and the apparent source motion E = E (H ) results in a amplied Doppler shift of the diraction images. Expanding in terms of we have 3 2 E = 1 2 cos H + ( + cos H ) 2 3 cos H 3 + O( 4 ) 0 . 2 Considering only the rst order terms in , Yn reads n d0 0 [1 2 cos H ] . Yn = Now, if we denominate Yn = n d0 0 /, the relative variation of Yn can be found as Yn Yn Yn = = 2 cos H . Yn Yn (18) (17) (16)

Thus the EGT scenario for the relative variation (shift) Yn /Yn of a diraction image predicts an amplitude twice bigger than the dipole speed of light anisotropy and diers from the SRT prediction where Yn /Yn = 0. We have compared these two predictions with experimental data obtained from measurements on light diraction produced by a laser beam in perpendicular incidence on a diraction grating as a function of the laser beam alignment relative to the Earths velocity vector. So far, we have three complete sets of independent observations made in three days, February 20 and 21 and March 9 of 2006. In order to avoid contaminations on the diraction observations due to atmospheric uctuations such as pressure and temperature among others, each set of observations were obtained in ten minutes by photographing the diraction image in 19 dierent positions of the angle H , from H = 00 up to H = 1800 in steps of 100 . A digital machine of 3.2 Mega pixels, xed in the telescope has been used. An o-line analysis of the digital images has been done, by using a special software. The method permits us to count the number of pixels between two points of the image. Our goal is to obtain only the relative variations in the position of the diraction images, and it is not necessary to do calibrations. We present here the measurements made on March 9, 2006 where a diraction grating of 600 lines/mm and a Helio-Neon Laser of 632.8 nm and 1.2mW has been used. Each position Yn has been measured with an accuracy of 99.87%. Fig.5 shows the amplied Doppler shift observed through the overlap of diraction images as a function of the laser beam alignment relative to the Earths velocity vector The images were ltered and amplied using always the same criteria. The colors in the images are articial and the vertical lines were placed only as a visual guide. A quantitative result, showing the relative amplied Doppler shift as function of the laser beam alignment, is summarized in Fig.6, where the SRT and EGT predictions are compares with the experimental data.
4. CONCLUSIONS AND REMARKS

An experimental and theoretical survey has been made on the one-way light propagation on Earth. The measurements have been obtained by using the TUPI diraction laser experiment, and they have been analyzed on the basis

5 of the EGT formalism. We nd that there is evidences for an anisotropic light propagation (a rst-order ether drift eect), because an amplied Doppler shift has been observed through the overlap of diraction images as a function of the laser beam alignment relative to the Earths velocity vector. This behavior arises of the Earths motion (solar barycenter) relative to the CMBR rest frame. These experimental results contradict the so called Lorentzs theorem [18], which states that the course of the relative rays is not aected by the motion of the Earth if quantities of the second order are neglected. Now, after our experimental results, that certainly will be conrmed by other similar experiments, we conclude that the course of the rays is aected by the motion of the Earth, and a predominant quantity of rst order describes well the experimental results. Of course, the observation of this eect requires a careful determination of the ether drift direction, and only in 1991 after the publication of the COBEs results, that we know this direction with accuracy. Our results point out that is not possible to neglect the preferred frame imposed by the cosmology. These results are well described by the Ether Gauge Theory (EGT) and which incorporates the cosmological time constraint. However, when only local aspects are considered the EGT predictions agree with the SRT predictions, they are basically indistinguishable locally. Finally, we are conscious that only after an independent conrmation of our experimental data that they will have a total credibility and we hope that this happens soon, because the experiment is easy to reproduce. Until then, we have the conviction that the universe behaves as a birefringent crystal ball.
5. ACKNOWLEDMENTS

This paper is a memorial tribute to our professor and friend C.M.G. Lattes, he was who introduced us the nonconventional aspects about relativity. We are very grateful to K.H. Tsui for his assistance in preparation of the manuscript.

[1] Roberts T. What is the Experimental basis of Special Relativity?. (http://math.ucr.edu/home/baez/physics/Relativity/SR/experiments.html, 2000). [2] Nodland B and Ralston J. P. Indication of Anisotropy in Electromagnetic Propagation over Cosmological Distances. Phys. Rev. Lett. 78, 3043-3046 (1997). [3] Smoot G. F. et al. Preliminary Results from the COBE Dierential Microwave Radiometers: Large angular Scale Isotropy of the Cosmic Microwave Background. Astrophysical Journal, 371, L1-L5 (1991). [4] Yaes R. J. Reconciling COBE Data with Relativity. Physics Today, March, 13 (1993). [5] Hans. S. General Relativity: An Introduction to the Theory of the Gravitational Field., 239 (Cambridge University Press, Cambridge, 1982). [6] Walker E. H. The cosmic Microwave Background Radiation As A Marker Of An Absolute Reference Frame. (http://users.rcn.com/wcri/wcri/absolute%20frame%20text.htm). [7] Navia. C. E., Augusto C. R. A., Robba M. B., Malheiro M. and Shigueoka H. Is There an Enhancement of Muons at sea level From transient Events?. Astrophysical Journal, 621: 1137-1145 (2005). [8] Lattes C. M. G. Private communication (1980). [9] Brans C. H. and Stewart D. R. Unaccelerated-Returning-Twin Paradox in Flat Space-Time. Phys. Rev. D8, 1662-1666 (1973). [10] Hatch R. R. In Search of an Ether Drift. Galilean Electrodynamics, Vol.13, 3-8 (2002). [11] Thangherlini F. R. Nuovo Cim. Suppl. Vol.20, 351 (1961). [12] Selleri F. Noninvariant One-way Velocity of Light. Founadtions of Physics, Vol.26, 641-664 (1996). [13] Augusto C. R. A. Navia C. E. and Robba M. B. Search for muons enhancement at sea level from transient solar activity. Phys. Rev. D71, 103011 (2005). [14] Navia C. E., Augusto C. R. A., Tsui K. H. and Robba M. B. Mini-Forbush events on the muon ux at sea level. Phys. Rev 72, 103001 (2005). [15] Smoot G. F. and Scoot D. Cosmic Background Radiation. European. Phys. J. N.1-4, 145-149 (2000). [16] Mller H., Hermann S., Braxmaier C., Schillr S. and Peters A. Modern Michelson-Morley Experiment using Cryogenic u Optical Resonators. Phys. Rev. Lett. 91, 0204001 (2003). [17] Ives H. E. and Stilwell G. R. An Experimental Study of the Rate of a Moving Atomic Clock. J. Opt. Soc. Am. 28, 215-226 (1938). [18] Lorentz H. A. The Theory of electrons:And its Applications to the Phenomena of light and Radiant Head. (Dover Phoenix Editions 2004).

FIG. 1: Space-time diagram (light cone) in the E frame according to the EGT formalism. It is possible to see that the axis-X (direction of VE ) behaves as a preferred direction. The slope of the lines are proportional to the inverse value of the speed of light. Ve is the resulting vector speed of light (emission vector). For illustrative purpose, a large value for (= 0.3) in this diagram has been used. The eect for instance, due to the Earth motion relative to the CMBR rest frame is tiny ( = 0.001237).
0.005 0.004 0.003 0.002 0.001

EGT ( = 0.001237 ) ERT

cE / cE

0.000 -0.001 -0.002 -0.003 -0.004 -0.005 0 20 40 60 80 100 120 140 160 180

H (degree)

FIG. 2: Angular dependence relative to the Earths velocity vector of the dipole speed of light. The lines represent the expected values according to the EGT ( = 0.001237) and SRT formalisms.

FIG. 3: The equatorial ensemble of the TUPI muon telescope (raster scan system). The axis of the telescope can be pointing for a pre-established direction. The diraction device is mounted in the shell of the telescope. The laser beam is parallel to the telescope axis. For H = 0 the telescope points to the Crater constellation.

FIG. 4: Schematic representation of the TUPI laser diraction experiment in the Earth (E frame). The emission vector is represented as Ve .

FIG. 5: Amplied Doppler shift observed through the overlap of diraction images as a function of the laser beam alignment relative to the Earths velocity vector. The images were ltered and amplied using always the same criteria. The colors in the images are articial and the vertical lines were placed only as a visual guide.

0.008 0.007 0.006 0.005 0.004 0.003 0.002 0.001 0.000 -0.001 -0.002 -0.003 -0.004 -0.005 -0.006 -0.007 -0.008 0

TUPI data EGT SRT

Yn / Yn

20

40

60

80

100 120 140 160 180

H (degree)
FIG. 6: Angular dependence relative to the earth vector velocity of the relative amplied Doppler shift of the diraction images. The expected curves according to the SRT and EGT( = 0.001237) and experimental data are shown.